Probabilistic Forecasting of Ground Magnetic Perturbation Spikes at Mid‐Latitude Stations

Abstract

The prediction of large fluctuations in the ground magnetic field (dB/dt) is essential for preventing damage from Geomagnetically Induced Currents. Directly forecasting these fluctuations has proven difficult, but accurately determining the risk of extreme events can allow for the worst of the damage to be prevented. Here we trained Convolutional Neural Network models for eight mid‐latitude magnetometers to predict the probability that dB/dt will exceed the 99th percentile threshold 30–60 min in the future. Two model frameworks were compared, a model trained using solar wind data from the Advanced Composition Explorer (ACE) satellite, and another model trained on both ACE and SuperMAG ground magnetometer data. The models were compared to examine if the addition of current ground magnetometer data significantly improved the forecasts of dB/dt in the future prediction window. A bootstrapping method was employed using a random split of the training and validation data to provide a measure of uncertainty in model predictions. The models were evaluated on the ground truth data during eight geomagnetic storms and a suite of evaluation metrics are presented. The models were also compared to a persistence model to ensure that the model using both datasets did not over‐rely on dB/dt values in making its predictions. Overall, we find that the models using both the solar wind and ground magnetometer data had better metric scores than the solar wind only and persistence models, and was able to capture more spatially localized variations in the dB/dt threshold crossings.

Sudden Commencements and Geomagnetically Induced Currents in New Zealand: Correlations and Dependance

“…These results demonstrate that the full vector, sub‐minute resolution magnetic field signature is important to consider when interpreting the space weather impact of a given event. Much work in recent times has focused on forecasting the one minute rate of change of the geomagnetic field (e.g., Blandin et al., 2022; Keesee et al., 2020; Pinto et al., 2022; Wintoft et al., 2015), or when it will exceed predefined thresholds (Camporeale et al., 2020; Coughlan et al., 2023; Pulkkinen et al., 2013; Smith, Forsyth, Rae, Garton, et al., 2021). However, we have shown that even if the magnitude of H ′ is predicted perfectly, and even though H ′ and GIC linearly correlate rather well (Mac Manus et al., 2017; Viljanen et al., 2001), any GIC derived through a simple correlation will still come with considerable uncertainty due to the orientation of H ′ and the sub‐minute frequency content of the magnetic changes.…”

“…However, in general it has often been assumed that a larger rate of change of the magnetic field will drive larger GICs (Mac Manus et al., 2017; Smith et al., 2022; Viljanen et al., 2001). For this reason, much recent effort has been made to forecast the rate of change of the magnetic field (e.g., Blandin et al., 2022; Keesee et al., 2020; Madsen et al., 2022; Pinto et al., 2022; Upendran et al., 2022; Wintoft et al., 2015), or the probability that it will exceed defined thresholds (e.g., Camporeale et al., 2020; Coughlan et al., 2023; Pulkkinen et al., 2013; Smith, Forsyth, Rae, Garton, et al., 2021). The focus on the magnetic field, rather than GICs, has partly been necessitated by the typical scarcity of freely available GIC observations, compared to the relative abundance of magnetic field measurements.…”